It has been found that inbred mice and rats can be immunized phrophylactically against tumors derived from mice and rats of the same genetic background (Gross (1943) Cancer Res. 3:323-326; Prehn et al. (1957) J. Natl. Cancer Inst. 18:769-778; Klein et al. (1960) Cancer Res. 20:1561-1572; Old et al. (1962) Ann NY Acad. Sci. 101:80-106; for review, see Srivastava et al. (1988) Immunology Today 9:78-83). These studies not only showed that mice vaccinated with inactivated cancer cells become immunized against subsequent challenges of live cancer cells but also demonstrated the existence of tumor-specific antigens.
Further studies revealed that the phenomenon of prophylactically induced immunity is tumor-specific. Although mice can be specifically immunized against the tumor cells that were used to immunize them they still remain sensitive to challenges with other unrelated tumors (Basombrio (1970) Cancer Res. 30:2458-2462, Globerson et al. (1964) J. Natl. Cancer Inst. 32:1229-1243). The demonstration of immunogenicity of cancer cells led to a search for the cancer-derived molecules which elicit resistance to tumor challenges. The general approach was to fractionate cancer cell-derived proteins and test them individually for their ability to immunize mice against the cancers from which the fractions were prepared (see Srivastava et al. (1988) supra; Old (1981) Cancer Res. 41:361-375). A number of proteins have been identified by this method, however, a large proportion of these proteins are related to a class of proteins known as stress-induced proteins or stress proteins (Lindquist et al. (1988) Annual Rev. Genet. 22:631-677). Because the stress proteins are among the most highly conserved and abundant proteins in nature, they are unlikely candidates for tumor specific antigens. Stress proteins have subsequently been shown to non covalently associate with a variety of peptides thereby to form stress protein-peptide complexes (Gething et al. (1992) Nature 355:33-45; Lindquist et al. (1988) supra; Young (1990) Annu. Rev. Immunol. 8:401-420; Flynn et al. 1991) Nature 353:726-730).
Studies have also shown that stress protein-peptide complexes lose their immunogenicity upon treatment with ATP (Udono et al. (1993) J. Exp. Med. 178:1391-1396). This treatment is known to dissociate the stress-protein peptide complex into its stress protein and peptide components. Considering that there are no differences in the structure of stress proteins derived from normal and tumor cells, and that stress proteins bind a wide spectrum of peptides in an ATP dependent manner it appears that the antigenicity of the stress protein-peptide complex results not from the stress protein per se, but from the peptide associated with the stress protein.
One of the major conceptual difficulties in cancer immunotherapy has been the possibility that human cancers, like cancers of experimental animals, are antigenically distinct. Clearly, there is some recent evidence for existence of common tumor antigens (Kawakami et al. (1992) J. Immunol. 148:638-643; Darrow et al. (1989) J. Immunol. 142:3329-3334), and this augurs well for prospects of cancer immunotherapy. Nonetheless, in light of the overwhelming evidence from experimental and human systems, it is reasonable to assume that at the very least, human tumors would show tremendous antigenic diversity and heterogeneity.
The prospect of identification of the immunogenic antigens of individual tumors from cancer patients (or even of `only` several different types of immunogenic antigens in case the antigens are shared), is daunting to the extent of being impractical. Conventional cancer therapies typically are based on the isolation and characterization of specific antigenic determinants which then may become the target for subsequent immunotherapies. In addition, although studies have demonstrated that mammals can be immunized prophylactically against tumors derived from mammals of the same genetic background, heretofore it has not been appreciated that a mammal harboring a tumor can be therapeutically immunized with a composition derived from its own tumor as a means of treating a cancer preexisting in the mammal.
Accordingly, it is an object of the instant invention to provide a novel method for therapeutically inhibiting proliferation of tumors in a mammal. The method described herein does not require the isolation and characterization of specific antigenic determinants, and accordingly provides a more rapid approach for making and using immunogenic compositions effective in inhibiting the proliferation of specific predetermined tumors in mammals.
This and other objects and features of the invention will be apparent from the description and claims which follow.